Videomics of the Upper Aero-Digestive Tract Cancer: Deep Learning Applied to White Light and Narrow Band Imaging for Automatic Segmentation of Endoscopic Images

An automated approach for real-time informative frames classification in laryngeal endoscopy using deep learning

PURPOSE

Informative image selection in laryngoscopy has the potential for improving automatic data extraction alone, for selective data storage and a faster review process, or in combination with other artificial intelligence (AI) detection or diagnosis models. This paper aims to demonstrate the feasibility of AI in providing automatic informative laryngoscopy frame selection also capable of working in real-time providing visual feedback to guide the otolaryngologist during the examination.

METHODS

Several deep learning models were trained and tested on an internal dataset (n = 5147 images) and then tested on an external test set (n = 646 images) composed of both white light and narrow band images. Four videos were used to assess the real-time performance of the best-performing model.

RESULTS

ResNet-50, pre-trained with the pretext strategy, reached a precision = 95% vs. 97%, recall = 97% vs, 89%, and the F1-score = 96% vs. 93% on the internal and external test set respectively (p = 0.062). The four testing videos are provided in the supplemental materials.

CONCLUSION

The deep learning model demonstrated excellent performance in identifying diagnostically relevant frames within laryngoscopic videos. With its solid accuracy and real-time capabilities, the system is promising for its development in a clinical setting, either autonomously for objective quality control or in conjunction with other algorithms within a comprehensive AI toolset aimed at enhancing tumor detection and diagnosis.

Laryngeal Cancer Screening During Flexible Video Laryngoscopy Using Large Computer Vision Models

OBJECTIVE

Develop an artificial intelligence assisted computer vision model to screen for laryngeal cancer during flexible laryngoscopy.

METHODS

Using laryngeal images and flexible laryngoscopy video recordings, we developed computer vision models to classify video frames for usability and cancer screening. A separate model segments any identified lesions on the frames. We used these computer vision models to construct a video stream annotation system. This system classifies findings from flexible laryngoscopy as "potentially malignant" or "probably benign" and segments any detected lesions. Additionally, the model provides a confidence level for each classification.

RESULTS

The overall accuracy of the flexible laryngoscopy cancer screening model was 92%. For cancer screening, it achieved a sensitivity of 97.7% and a specificity of 76.9%. The segmentation model attained an average precision at a 0.50 intersection-over-union of 0.595. The confidence level for positive screening results can assist clinicians in counseling patients regarding the findings.

CONCLUSION

Our model is highly sensitive and adequately specific for laryngeal cancer screening. Segmentation helps endoscopists identify and describe potential lesions. Further optimization is required to enable the model's deployment in clinical settings for real-time annotation during flexible laryngoscopy.

Computer Vision and Videomics in Otolaryngology-Head and Neck Surgery: Bridging the Gap Between Clinical Needs and the Promise of Artificial Intelligence

This article discusses the role of computer vision in otolaryngology, particularly through endoscopy and surgery. It covers recent applications of artificial intelligence (AI) in nonradiologic imaging within otolaryngology, noting the benefits and challenges, such as improving diagnostic accuracy and optimizing therapeutic outcomes, while also pointing out the necessity for enhanced data curation and standardized research methodologies to advance clinical applications. Technical aspects are also covered, providing a detailed view of the progression from manual feature extraction to more complex AI models, including convolutional neural networks and vision transformers and their potential application in clinical settings.

The Use of Deep Learning Software in the Detection of Voice Disorders: A Systematic Review

OBJECTIVE

To summarize the use of deep learning in the detection of voice disorders using acoustic and laryngoscopic input, compare specific neural networks in terms of accuracy, and assess their effectiveness compared to expert clinical visual examination.

DATA SOURCES

Embase, MEDLINE, and Cochrane Central.

REVIEW METHODS

Databases were screened through November 11, 2023 for relevant studies. The inclusion criteria required studies to utilize a specified deep learning method, use laryngoscopy or acoustic input, and measure accuracy of binary classification between healthy patients and those with voice disorders.

RESULTS

Thirty-four studies met the inclusion criteria, with 18 focusing on voice analysis, 15 on imaging analysis, and 1 both. Across the 18 acoustic studies, 21 programs were used for identification of organic and functional voice disorders. These technologies included 10 convolutional neural networks (CNNs), 6 multilayer perceptrons (MLPs), and 5 other neural networks. The binary classification systems yielded a mean accuracy of 89.0% overall, including 93.7% for MLP programs and 84.5% for CNNs. Among the 15 imaging analysis studies, a total of 23 programs were utilized, resulting in a mean accuracy of 91.3%. Specifically, the twenty CNNs achieved a mean accuracy of 92.6% compared to 83.0% for the 3 MLPs.

CONCLUSION

Deep learning models were shown to be highly accurate in the detection of voice pathology, with CNNs most effective for assessing laryngoscopy images and MLPs most effective for assessing acoustic input. While deep learning methods outperformed expert clinical exam in limited comparisons, further studies integrating external validation are necessary.

Multi-Instance Learning for Vocal Fold Leukoplakia Diagnosis Using White Light and Narrow-Band Imaging: A Multicenter Study

OBJECTIVES

Vocal fold leukoplakia (VFL) is a precancerous lesion of laryngeal cancer, and its endoscopic diagnosis poses challenges. We aim to develop an artificial intelligence (AI) model using white light imaging (WLI) and narrow-band imaging (NBI) to distinguish benign from malignant VFL.

METHODS

A total of 7057 images from 426 patients were used for model development and internal validation. Additionally, 1617 images from two other hospitals were used for model external validation. Modeling learning based on WLI and NBI modalities was conducted using deep learning combined with a multi-instance learning approach (MIL). Furthermore, 50 prospectively collected videos were used to evaluate real-time model performance. A human-machine comparison involving 100 patients and 12 laryngologists assessed the real-world effectiveness of the model.

RESULTS

The model achieved the highest area under the receiver operating characteristic curve (AUC) values of 0.868 and 0.884 in the internal and external validation sets, respectively. AUC in the video validation set was 0.825 (95% CI: 0.704-0.946). In the human-machine comparison, AI significantly improved AUC and accuracy for all laryngologists (p < 0.05). With the assistance of AI, the diagnostic abilities and consistency of all laryngologists improved.

CONCLUSIONS

Our multicenter study developed an effective AI model using MIL and fusion of WLI and NBI images for VFL diagnosis, particularly aiding junior laryngologists. However, further optimization and validation are necessary to fully assess its potential impact in clinical settings.

LEVEL OF EVIDENCE

3 Laryngoscope, 2024.

Exploring the landscape of AI-assisted decision-making in head and neck cancer treatment: a comparative analysis of NCCN guidelines and ChatGPT responses

PURPOSE

Recent breakthroughs in natural language processing and machine learning, exemplified by ChatGPT, have spurred a paradigm shift in healthcare. Released by OpenAI in November 2022, ChatGPT rapidly gained global attention. Trained on massive text datasets, this large language model holds immense potential to revolutionize healthcare. However, existing literature often overlooks the need for rigorous validation and real-world applicability.

METHODS

This head-to-head comparative study assesses ChatGPT's capabilities in providing therapeutic recommendations for head and neck cancers. Simulating every NCCN Guidelines scenarios. ChatGPT is queried on primary treatments, adjuvant treatment, and follow-up, with responses compared to the NCCN Guidelines. Performance metrics, including sensitivity, specificity, and F1 score, are employed for assessment.

RESULTS

The study includes 68 hypothetical cases and 204 clinical scenarios. ChatGPT exhibits promising capabilities in addressing NCCN-related queries, achieving high sensitivity and overall accuracy across primary treatment, adjuvant treatment, and follow-up. The study's metrics showcase robustness in providing relevant suggestions. However, a few inaccuracies are noted, especially in primary treatment scenarios.

CONCLUSION

Our study highlights the proficiency of ChatGPT in providing treatment suggestions. The model's alignment with the NCCN Guidelines sets the stage for a nuanced exploration of AI's evolving role in oncological decision support. However, challenges related to the interpretability of AI in clinical decision-making and the importance of clinicians understanding the underlying principles of AI models remain unexplored. As AI continues to advance, collaborative efforts between models and medical experts are deemed essential for unlocking new frontiers in personalized cancer care.

Semi-automatic standardized analysis method to objectively evaluate near-infrared fluorescent dyes in image-guided surgery

Significance

Near-infrared fluorescence imaging still lacks a standardized, objective method to evaluate fluorescent dye efficacy in oncological surgical applications. This results in difficulties in translation between preclinical to clinical studies with fluorescent dyes and in the reproduction of results between studies, which in turn hampers further clinical translation of novel fluorescent dyes.

Aim

Our aim is to develop and evaluate a semi-automatic standardized method to objectively assess fluorescent signals in resected tissue.

Approach

A standardized imaging procedure was designed and quantitative analysis methods were developed to evaluate non-targeted and tumor-targeted fluorescent dyes. The developed analysis methods included manual selection of region of interest (ROI) on white light images, automated fluorescence signal ROI selection, and automatic quantitative image analysis. The proposed analysis method was then compared with a conventional analysis method, where fluorescence signal ROIs were manually selected on fluorescence images. Dice similarity coefficients and intraclass correlation coefficients were calculated to determine the inter- and intraobserver variabilities of the ROI selections and the determined signal- and tumor-to-background ratios.

Results

The proposed non-targeted fluorescent dyes analysis method showed statistically significantly improved variabilities after application on indocyanine green specimens. For specimens with the targeted dye SGM-101, the variability of the background ROI selection was statistically significantly improved by implementing the proposed method.

Conclusion

Semi-automatic methods for standardized quantitative analysis of fluorescence images were successfully developed and showed promising results to further improve the reproducibility and standardization of clinical studies evaluating fluorescent dyes.

The Detection of Nasopharyngeal Carcinomas Using a Neural Network Based on Nasopharyngoscopic Images

OBJECTIVE

To construct and validate a deep convolutional neural network (DCNN)-based artificial intelligence (AI) system for the detection of nasopharyngeal carcinoma (NPC) using archived nasopharyngoscopic images.

METHODS

We retrospectively collected 14107 nasopharyngoscopic images (7108 NPCs and 6999 noncancers) to construct a DCNN model and prepared a validation dataset containing 3501 images (1744 NPCs and 1757 noncancers) from a single center between January 2009 and December 2020. The DCNN model was established using the You Only Look Once (YOLOv5) architecture. Four otolaryngologists were asked to review the images of the validation set to benchmark the DCNN model performance.

RESULTS

The DCNN model analyzed the 3501 images in 69.35 s. For the validation dataset, the precision, recall, accuracy, and F1 score of the DCNN model in the detection of NPCs on white light imaging (WLI) and narrow band imaging (NBI) were 0.845 ± 0.038, 0.942 ± 0.021, 0.920 ± 0.024, and 0.890 ± 0.045, and 0.895 ± 0.045, 0.941 ± 0.018, and 0.975 ± 0.013, 0.918 ± 0.036, respectively. The diagnostic outcome of the DCNN model on WLI and NBI images was significantly higher than that of two junior otolaryngologists (p < 0.05).

CONCLUSION

The DCNN model showed better diagnostic outcomes for NPCs than those of junior otolaryngologists. Therefore, it could assist them in improving their diagnostic level and reducing missed diagnoses.

LEVEL OF EVIDENCE

3 Laryngoscope, 134:127-135, 2024.

Deep Learning Techniques and Imaging in Otorhinolaryngology-A State-of-the-Art Review

Over the last decades, the field of medicine has witnessed significant progress in artificial intelligence (AI), the Internet of Medical Things (IoMT), and deep learning (DL) systems. Otorhinolaryngology, and imaging in its various subspecialties, has not remained untouched by this transformative trend. As the medical landscape evolves, the integration of these technologies becomes imperative in augmenting patient care, fostering innovation, and actively participating in the ever-evolving synergy between computer vision techniques in otorhinolaryngology and AI. To that end, we conducted a thorough search on MEDLINE for papers published until June 2023, utilizing the keywords 'otorhinolaryngology', 'imaging', 'computer vision', 'artificial intelligence', and 'deep learning', and at the same time conducted manual searching in the references section of the articles included in our manuscript. Our search culminated in the retrieval of 121 related articles, which were subsequently subdivided into the following categories: imaging in head and neck, otology, and rhinology. Our objective is to provide a comprehensive introduction to this burgeoning field, tailored for both experienced specialists and aspiring residents in the domain of deep learning algorithms in imaging techniques in otorhinolaryngology.

Artificial Intelligence for Upper Aerodigestive Tract Endoscopy and Laryngoscopy: A Guide for Physicians and State-of-the-Art Review

OBJECTIVE

The endoscopic and laryngoscopic examination is paramount for laryngeal, oropharyngeal, nasopharyngeal, nasal, and oral cavity benign lesions and cancer evaluation. Nevertheless, upper aerodigestive tract (UADT) endoscopy is intrinsically operator-dependent and lacks objective quality standards. At present, there has been an increased interest in artificial intelligence (AI) applications in this area to support physicians during the examination, thus enhancing diagnostic performances. The relative novelty of this research field poses a challenge both for the reviewers and readers as clinicians often lack a specific technical background.

DATA SOURCES

Four bibliographic databases were searched: PubMed, EMBASE, Cochrane, and Google Scholar.

REVIEW METHODS

A structured review of the current literature (up to September 2022) was performed. Search terms related to topics of AI, machine learning (ML), and deep learning (DL) in UADT endoscopy and laryngoscopy were identified and queried by 3 independent reviewers. Citations of selected studies were also evaluated to ensure comprehensiveness.

CONCLUSIONS

Forty-one studies were included in the review. AI and computer vision techniques were used to achieve 3 fundamental tasks in this field: classification, detection, and segmentation. All papers were summarized and reviewed.

IMPLICATIONS FOR PRACTICE

This article comprehensively reviews the latest developments in the application of ML and DL in UADT endoscopy and laryngoscopy, as well as their future clinical implications. The technical basis of AI is also explained, providing guidance for nonexpert readers to allow critical appraisal of the evaluation metrics and the most relevant quality requirements.

Instance segmentation of upper aerodigestive tract cancer: site-specific outcomes

Objective

To achieve instance segmentation of upper aerodigestive tract (UADT) neoplasms using a deep learning (DL) algorithm, and to identify differences in its diagnostic performance in three different sites: larynx/hypopharynx, oral cavity and oropharynx.

Methods

A total of 1034 endoscopic images from 323 patients were examined under narrow band imaging (NBI). The Mask R-CNN algorithm was used for the analysis. The dataset split was: 935 training, 48 validation and 51 testing images. Dice Similarity Coefficient (Dsc) was the main outcome measure.

Results

Instance segmentation was effective in 76.5% of images. The mean Dsc was 0.90 ± 0.05. The algorithm correctly predicted 77.8%, 86.7% and 55.5% of lesions in the larynx/hypopharynx, oral cavity, and oropharynx, respectively. The mean Dsc was 0.90 ± 0.05 for the larynx/hypopharynx, 0.60 ± 0.26 for the oral cavity, and 0.81 ± 0.30 for the oropharynx. The analysis showed inferior diagnostic results in the oral cavity compared with the larynx/hypopharynx (p < 0.001).

Conclusions

The study confirms the feasibility of instance segmentation of UADT using DL algorithms and shows inferior diagnostic results in the oral cavity compared with other anatomic areas.

The tumor ecosystem in head and neck squamous cell carcinoma and advances in ecotherapy

The development of head and neck squamous cell carcinoma (HNSCC) is a multi-step process, and its survival depends on a complex tumor ecosystem, which not only promotes tumor growth but also helps to protect tumor cells from immune surveillance. With the advances of existing technologies and emerging models for ecosystem research, the evidence for cell-cell interplay is increasing. Herein, we discuss the recent advances in understanding the interaction between tumor cells, the major components of the HNSCC tumor ecosystem, and summarize the mechanisms of how biological and abiotic factors affect the tumor ecosystem. In addition, we review the emerging ecological treatment strategy for HNSCC based on existing studies.

The Role of Peritumoral Depapillation and Its Impact on Narrow-Band Imaging in Oral Tongue Squamous Cell Carcinoma

A recent study reported that the occurrence of depapillated mucosa surrounding oral tongue squamous cell carcinomas (OTSCC) is associated with perineural invasion (PNI). The present study evaluates the reliability of depapillation as a PNI predictor and how it could affect narrow-band imaging (NBI) performance. This is thus a retrospective study on patients affected by OTSCC submitted to radical surgery. The preoperative endoscopy was evaluated to identify the presence of depapillation. Differences in distribution between depapillation and clinicopathological variables were analyzed. NBI vascular patterns were reported, and the impact of depapillation on those was studied. We enrolled seventy-six patients. After evaluation of the preoperative endoscopies, 40 (53%) patients had peritumoral depapillation, while 59 (78%) had a positive NBI pattern. Depapillation was strongly correlated to PNI, 54% vs. 28% (p = 0.022). Regarding the NBI pattern, there was no particular association with depapillation-associated tumors. The presence of depapillation did not affect the intralesional pattern detected by the NBI, while no NBI-positive pattern was found in the depapillation area. Finally, the NBI-guided resection margins were not affected by depapillation. Peritumoral depapillation is a reliable feature for PNI in OTSCC. NBI margin detection is not impaired by depapillation.

Intraoperative Imaging Techniques to Improve Surgical Resection Margins of Oropharyngeal Squamous Cell Cancer: A Comprehensive Review of Current Literature

Inadequate resection margins in head and neck squamous cell carcinoma surgery necessitate adjuvant therapies such as re-resection and radiotherapy with or without chemotherapy and imply increasing morbidity and worse prognosis. On the other hand, taking larger margins by extending the resection also leads to avoidable increased morbidity. Oropharyngeal squamous cell carcinomas (OPSCCs) are often difficult to access; resections are limited by anatomy and functionality and thus carry an increased risk for close or positive margins. Therefore, there is a need to improve intraoperative assessment of resection margins. Several intraoperative techniques are available, but these often lead to prolonged operative time and are only suitable for a subgroup of patients. In recent years, new diagnostic tools have been the subject of investigation. This study reviews the available literature on intraoperative techniques to improve resection margins for OPSCCs. A literature search was performed in Embase, PubMed, and Cochrane. Narrow band imaging (NBI), high-resolution microendoscopic imaging, confocal laser endomicroscopy, frozen section analysis (FSA), ultrasound (US), computed tomography scan (CT), (auto) fluorescence imaging (FI), and augmented reality (AR) have all been used for OPSCC. NBI, FSA, and US are most commonly used and increase the rate of negative margins. Other techniques will become available in the future, of which fluorescence imaging has high potential for use with OPSCC.

Accuracy of narrow band imaging for detecting the malignant transformation of oral potentially malignant disorders: A systematic review and meta-analysis

Objective

Oral potentially malignant disorders (OPMDs) are a spectrum of diseases that harbor the potential of malignant transformation and developing into oral squamous cell carcinoma (OSCC). Narrow band imaging (NBI) has been clinically utilized for the adjuvant diagnosis of OPMD and OSCC. This study aimed to comprehensively evaluate the diagnostic accuracy of NBI for malignant transformations of OPMD by applying the intraepithelial papillary capillary loop (IPCL) classification approach.

Methods

Studies reporting the diagnostic validity of NBI in the detection of OPMD/OSCC were selected. Four databases were searched and 11 articles were included in the meta-analysis. We performed four subgroup analyses by defining IPCL I/II as negative diagnostic results and no/mild dysplasia as negative pathological outcome. Pooled data were analyzed using random-effects models. Meta-regression analysis was performed to explore heterogeneity.

Results

After pooled analysis of the four subgroups, we found that subgroup 1, defining IPCL II and above as a clinically positive result, demonstrated the most optimal overall diagnostic accuracy for the malignant transformation of OPMDs, with a sensitivity and specificity of NBI of 0.87 (95% confidence interval (CI) [0.67, 0.96], p < 0.001) and 0.83 [95% CI (0.56, 0.95), p < 0.001], respectively; while the other 3 subgroups displayed relatively low sensitivity or specificity.

Conclusions

NBI is a promising and non-invasive adjunctive tool for identifying malignant transformations of OPMDs. The IPCL grading is currently a sound criterion for the clinical application of NBI. After excluding potentially false positive results, these oral lesions classified as IPCL II or above are suggested to undergo biopsy for early and accurate diagnosis as well as management.